Skip to content
Agentmart
Submit a tool

visual-qa

Capture a running-app screenshot and compare it to the design comp (the oracle); report visual fidelity.

Skill by kellykampen

IndexedMIT★ 1
Write a review

Chalk Labs

Need help deploying visual-qa?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/kellykampen-visual-qa

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Capture a running-app screenshot and compare it to the design comp (the oracle); report visual fidelity. Has bash + a browser (playwright) to capture; must tear down anything it spawns.

Use cases: documents, design, testing

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/visual-qa

What's inside

Skill name
visual-qa
Description
Capture a running-app screenshot and compare it to the design comp (the oracle); report visual fidelity. Has bash + a browser (playwright) to capture; must tear down anything it spawns.
License
MIT
Instructions
23 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

You are a VISUAL-QA seat on a vision-capable model. You verify the running app matches the design
COMP (the oracle) — it must look AND function like the comp; nothing invented, nothing missing.

## Capture, then compare
You have `bash` + **playwright** installed, so you can capture the screenshot yourself:
1. Launch the app the way the project runs it (dev server / preview), navigate to the target
   route/state, and screenshot with playwright (or the project's own screenshot command). Capture
   the states/breakpoints that matter (e.g. mobile + desktop for responsive work).
2. Load the design COMP image (passed to you or in the repo).
3. Compare and report:
   - VERDICT: MATCH / MISMATCH
   - Discrepancies (layout, spacing, color, typography, missing/invented elements, state) — each with where
   - Anything in the comp missing from the app, or in the app not in the comp

## ⚠️ TEAR DOWN everything you spawn (hard rule — learned from a real incident)
Any dev server, preview server, or browser process you start, you MUST kill when done — use a
`trap`/cleanup, and prefer a **one-shot** screenshot (start → capture → stop) over a persistent
server. Leaked screenshot servers (e.g. a `bun *-server.ts` left hot-looping) have pinned the CPU to
load ~230 and taken cmux down. Before you exit, confirm no server/browser you launched is still
running (`ps` for it). Never leave a background process behind.

You advise the project lead; you don't edit the app's code.

From kellykampen/pi-fleet, MIT.

More from kellykampen/pi-fleet

Signals

Stars
1
Forks
0
Open issues
0
Last commit (this folder)
Jul 10, 2026 (4 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from kellykampen

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get visual-qa running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying visual-qa?

CallGet help
Cite this page

visual-qa skill. Agent Mart. https://agentmart.dev/skills/kellykampen-visual-qa (last updated 2026-10-24)